i
UNIVERSIDADE ESTADUAL DE CAMPINAS
Atividade de forrageamento e comportamento alimentar de duas
espécies sintópicas de Mullidae (Perciformes) no Arquipélago de
Fernando de Noronha, Pernambuco
Aluno: João Paulo Krajewski
Orientador: Ivan Sazima
Dissertação apresentada ao Instituto de Biologia da
Universidade Estadual de Campinas para a
obtenção do título de Mestre em Ecologia.
Campinas – 2005
iii
BANCA EXAMINADORA
Prof. Dr. Ivan Sazima
Prof. Dr. André Victor Lucci Freitas
Prof. Dra. Fosca Pedini Pereira Leite
Prof. Dra. Virginia Sanches Uieda
iv
“Bons naturalistas têm que ser historiadores”.
Gould, S. J.
v
AGRADECIMENTOS
Cristina e Ivan e Roberta, pela amizade, ajuda e tudo que ensinaram durante os
inesquecíveis anos de convivência.
A Roberta, pelo amor, paciência e ajuda (até debaixo d’água!). Por ser como é!
A André, Ângela e Ricardo, minha querida família, por toda ajuda e, principalmente,
por sempre me apoiarem e admirarem o que faço!
A meus queridos avós (Cordélia e Maury), que mesmo não sabendo nadar
“viajaram” comigo ao mundo dos peixes recifais! Obrigado pelo carinho e ajuda de sempre.
A minha família ituana (Angélica, Bonaldo, Bruno, Mário e Pedro) pelo interesse e
respeito pelo que faço. Ao Marinho e Yara pela grande admiração por meu trabalho e por
tudo que é vivo!
A Paulo Roberto Guimarães Jr (Miúdo), pela gentil e indispensável ajuda com as
análises estatísticas e indispensáveis sugestões durante a preparação do exame de
qualificação. À querida Marlies Sazima, pelo carinho e grande ajuda na identificação das
algas de Noronha, além de muitos outros favores!
A todas as pessoas e instituições que me ajudaram em Fernando de Noronha. Ao
Projeto Golfinho Rotador, especialmente José Martins da Silva Jr, cuja ajuda na logística
e atividades foi de grande importância para o desenvolvimento desta tese e de outros
estudos em Noronha. IBAMA, principalmente Marco Aurélio Silva, cujo apoio logístico e
permissão para a realização deste estudo no Arquipélago foi fundamental. Ao João Flores,
do IBAMA, por tentar ressuscitar nosso motor de popa (tantas vezes...). Alice Grossmann,
do Projeto Tamar, pela companhia agradável e ajuda no trabalho de campo. Operadoras de
vi
mergulho de Fernando de Noronha, Águas Claras, Atlantis e Noronha Divers, pelo
auxílio com equipamentos e saídas de mergulho.
A equipe do Terra da Gente (Raul, Toni, Liana e Ciro) pelas lindas matérias sobre
nossos trabalhos.
À CAPES, CNPq, FAPESP e PROAP-Unicamp pela contribuição financeira para o
desenvolvimento do projeto.
A todos os membros da banca e pré-banca, André Victor L Freitas, Sérgio R
Floeter, Fosca P P Leite, Virgínia S Uieda , pela atenção, gentileza e comentários.
A Carlos Eduardo L Ferreira e Sérgio R Floeter, pelas interessantes discussões
sobre peixes recifais.
As pessoas queridas que preferem permanecer anônimas, mas cuja a ajuda e carinho
foram indispensáveis!
vii
Índice
Resumo 1
Abstract 2
Introdução geral 3
Objetivos 11
Referências 12
Capítulo 1: Foraging activity and behavior of two syntopic goatfish species (Perciformes:
Mullidae) in Fernando de Noronha Archipelago, Equatorial SW Atlantic.
(Manuscrito submetido para publicação em Journal of Fish Biology)
1.1: Abstract 15
1.2: Key words 15
1.3: Introduction 16
1.4: Materials and Methods 17
1.5: Results 19
1.6: Discussion 23
1.7: Acknowledgments 27
1.8: References 28
Capítulo 2: Plankton-picking by the goatfish Pseudupeneus maculatus (Mullidae), a
specialized bottom forager
(Manuscrito aceito para publicação em Journal of Fish Biology)
2.1: Abstract 31
2.2: Key words 31
viii
2.3: Full text 31
2.4: Aknoledgements 37
2.5: References 38
Capítulo 3: The association of the goatfish Mulloidichthys martinicus with the grunt
Haemulon chrysargyreum: an example of protective mimicry
(Manuscrito publicado em Biota Neotropica, Volume 4(2))
3.1: Abstract 42
3.2: Key words 42
3.3: Resumo 43
3.4: Palavras chave 43
3.5: Introduction 44
3.6: Materials and Methods 45
3.7: Results 45
3.8: Discussion 48
3.9: Acknowledgments 49
3.10: References 50
Discussão geral 52
Referências bibliográficas 54
ix
Índice de Figuras
Introdução geral
Figura 1. Pseudupeneus maculatus forrageia revolvendo e
fossando o substrato, assim atraindo atenção de predadores
oportunistas como o budião (Halichoeres poeyi), que se alimenta
das presas não apanhadas pelo mulídeo. (Foto: Roberta Martini
Bonaldo).
4
Figura 2. Localização do Arquipélago de Fernando de Noronha
no Atlântico Ocidental (a) e mapa do arquipélago (b). A principal
área de estudo, Praia da Conceição, está marcada com “*” no
mapa do arquipélago.
7
Figura 3. Costão rochoso da Praia da Conceição, Fernando de
Noronha, principal área de estudo.
7
Figura 4. Os quatro tipos de substratos estudados na Paia da
Conceição: algas pardas (b), misto (a), arenoso (c) e rochoso (d).
8
Figura 5. Mulloidichthys martinicus (a) e Pseudupeneus
maculatus (b) têm ampla distribuição no Atlântico Ocidental,
sendo sintópicos em diversos locais de Fernando de Noronha.
9
Capítulo 1
Figura 1. Frequency of feeding events of Mulloidichthys
martinicus and Pseudupeneus maculatus on each substrate type,
and the actual relative abundances of those substrates at
Fernando de Noronha Archipelago.
20
x
Figura 2. A hypothetic goatfish displaying the six feeding modes
(from the left to the right) recorded for Mulloidichthys martinicus
and Pseudupeneus maculatus at Fernando de Noronha
Archipelago. The former species displayed only excavate, push,
drift and bite, whereas the latter one displayed all modes. Based
on video frames and photographs.
22
Figura 3. Frequency of each feeding mode by Mulloidichthys
martinicus and Pseudupeneus maculatus at Fernando de Noronha
Archipelago.
23
Capítulo 2 Figura 1. A group of juvenile, recently settled goatfish,
Pseudupeneus maculatus, rising in the water column to feed on
plankton. The white spots in the water column are due to the
flash light reflected on the zooplankton.
33
Figura 2. A mixed group of the goatfish Pseudupeneus
maculatus and the damselfish Abudefduf saxatilis. The goatfish
are feeding on drifting crustaceans, close to the bottom. Note the
protruded mouth of the individual on the left side of the picture
(taken from a video frame).
34
Capítulo 3 Figura 1. The yellow goatfish (Mulloidichthys martinicus) is
difficult to tell apart from the smallmouth grunt (Haemulon
chrysargyreum) while in mixed schools and viewed under
natural light.
47
Figura 2. The yellow goatfish adopts similar posture and
behaviour while in a mixed school with the smallmouth grunt.
Note overall resemblance in shape and pattern between the two
species even under artificial light.
47
xi
Figura 3. A previously inactive group of yellow goatfish (lower
fish) joins a smallmouth grunt school when chased by a would be
predator (in this case a diver).
48
xii
Índice de Tabelas
Tabela I. Comparisons between Mulloidichthys martinicus and
Pseudupeneus maculatus at Fernando de Noronha Archipelago:
Student t test for feeding rates and distance roamed per time and
Mann-Whitney for time spent during each feeding event. The
parentheses indicate the total number of bouts for feeding rate
and distance roamed per time data and number of individuals
for time spent in each feeding event data.
21
1
Resumo
A família Mullidae (Perciformes) inclui cerca de 50 espécies de peixes distribuídas em
áreas tropicais e subtropicais dos Oceanos Índico, Pacífico e Atlântico. Os mulídeos
forrageiam no substrato, orientados principalmente por um par de barbilhões. Foi aqui
estudada, comparativamente, a atividade de forrageamento e o comportamento alimentar de
Mulloidichthys martinicus e Pseudupeneus maculatus, que vivem em sintopia no
Arquipélago de Fernando de Noronha (Pernambuco). Foi também registrado o
comportamento de forrageamento de grupos de P. maculatus na Reserva Biológica Marinha
do Arvoredo (Santa Catarina) e a formação de cardumes mistos sem atividade alimentar por
M. martinicus em Fernando de Noronha. As duas espécies diferiram em todos os aspectos
analisados. Pseudupeneus maculatus forrageia sobre três tipos de substrato, preferindo
substrato misto composto por areia e algas. Mulloidichthys martinicus explora dois tipos de
substrato, preferindo substrato arenoso. Pseudupeneus maculatus apresenta menor
freqüência alimentar, percorrendo menor distância por tempo durante o forrageamento e
apresenta repertório comportamental diversificado ao forragear. Adicionalmente, foram
registrados indivíduos de P. maculatus alimentando-se de plâncton na coluna d’água
quando forrageando em grupos, o primeiro registro de planctofagia para a espécie. Foi
também estudado um possível caso de mimetismo de proteção entre M. martinicus e
Haemulon chrysargyreum (Haemulidae), que freqüentemente formam cardumes mistos em
Fernando de Noronha. Pseudupeneus maculatus é aqui caracterizada como mais versátil, na
seleção de substrato para forrageio e em seu comportamento alimentar, quando comparada
a M. martinicus. Apesar de apresentarem diversas características em comum, as espécies de
Mullidae geralmente diferem em sua atividade de forrageamento, indicando que sua
simples caracterização como fossadores generalizados de substratos não consolidados é
inadequada.
2
Abstract
The goatfishes (Mullidae), include about 50 species distributed in tropical and subtropical
Indo-Pacific e Atlantic Oceans. All goatfishes forage on the bottom, primarily oriented by a
pair of chemosensorial barbels. The foraging behaviour and activity of Mulloidichthys
martinicus e Pseudupeneus maculatus was studied comparatively at Fernando de Noronha
Archipelago two goatfishes syntopic in (Brazil, Pernambuco), where these two species are
syntopic. The behaviour of P. maculatus when foraging in groups at the Reserva Biológica
Marinha do Arvoredo (Brazil, Santa Catarina) and mixed schooling behaviour of M.
martinicus at Fernando de Noronha was also studied. The two species differed in all
analyzed aspects. Pseudupeneus maculatus forages over tree substrate types but prefers
mixed substrate composed of sand and algae. Mulloidichthys martinicus forages over two
substrate types, preferring sandy substrate. Pseudupeneus maculatus has a lower feeding
rate, roams less per given time and displays a more diverse feeding mode repertoire when
compared with M. martinicus. Moreover, a study of plankton-feeding behaviour by P.
maculatus when foraging in groups at Fernando de Noronha and the Reserva Biológica
Marinha do Arvoredo was made, the first record of plankton-feeding for P. maculatus. A
presumed example of protective mimicry by M. martinicus and Haemulon chrysargyreum
(Haemulidae), which frequently form mixed schools in Fernando de Noronha, was also
studied. Pseudupeneus maculatus is here considered as a more versatile bottom forager
when compared to M. martinicus. Notwithstanding the overall similarity between mullid
species, they do differ in their substrate preferences and foraging activity, which indicates
that goatfishes can not be characterized simply as generalized soft bottom foragers.
3
INTRODUÇÃO GERAL
A família Mullidae
A família Mullidae (Perciformes) inclui cerca de 50 espécies de peixes de médio
porte (cerca de 10-30cm), distribuídas em áreas tropicais e subtropicais dos Oceanos
Índico, Pacífico e Atlântico (Froese & Pauly, 2005). As espécies desta família são
encontradas principalmente sobre fundos não-consolidados (lama, areia, cascalho), a
maioria delas próximas a recifes, em profundidades de cinco a 140 metros (Munro, 1976).
Durante a fase larval, os mulídeos são exclusivamente planctófagos, porém, após o
seu recrutamento no ambiente recifal, sofrem mudanças drásticas e relativamente rápidas,
passando a apresentar hábito bentônico. Dentre as mudanças sofridas durante a fase larval,
o desenvolvimento de um par de barbilhões quimio-sensoriais logo abaixo da boca é uma
das mais marcantes e está diretamente relacionada com o habito bentônico. O par de
barbilhões é usado na detecção, captura e manipulação de presas, possibilitando aos
mulídeos explorar eficientemente o substrato (Gosline, 1984; McCormick, 1995). Assim,
diferente de muitas outras famílias de peixes recifais, que têm representantes tanto
bentônicos como planctófagos, todos os mulídeos são carnívoros que forrageiam no
substrato (zoobentívoros) quando adultos (Gosline, 1984; McCormick, 1995).
Durante sua atividade alimentar, os mulídeos revolvem e fossam o substrato não
consolidado ao redor dos recifes (Gosline, 1984; McCormick, 1995), misturando o
sedimento e alterando a topografia e a composição da infauna (Barnes et al., 1993,
McCormick, 1995) (Figura 1). Durante essa perturbação do substrato, os mulídeos
desenterram e afugentam invertebrados e peixes bentônicos, atraindo a atenção de
predadores oportunistas. Estes predadores se associam momentaneamente aos mulídeos,
4
aproveitando os itens alimentares expostos por sua atividade aumentando, assim, seu
sucesso de caça (e.g. Aronson & Sanderson, 1987; Helfman et al., 1997) (Figura 1). Nesta
associação, os peixes seguidos (no caso, os mulídeos) são chamados de nucleares e os
peixes que se associam ao nuclear, de seguidores. Aparentemente, esta associação é
benéfica ao seguidor e geralmente inócua aos nucleares (Aronson & Sanderson, 1987).
Figura 1. Pseudupeneus maculatus forrageia revolvendo e fossando o substrato, assim
atraindo atenção de predadores oportunistas como o budião (Halichoeres poeyi), que se
alimenta das presas não apanhadas pelo mulídeo. (Foto: Roberta Martini Bonaldo).
As espécies de Mullidae em geral forrageiam sobre substratos não consolidados e as
sintópicas podem se sobrepor no uso de recursos alimentares (e.g. Munro, 1976;
McCormick, 1995). Entretanto, espécies sintópicas de Mullidae geralmente apresentam
diferentes modos de forragear e exploram diferentes microhábitats e tipos de substrato (e.g.
Hobson, 1974; Munro, 1976; McCormick, 1995). De modo geral, há espécies de Mullidae
associadas a fundos não consolidados (principalmente arenosos) e espécies associadas a
uma maior variedade de substratos, como fundos de algas, “entulho” (amontoados de corais
5
e conchas danificadas), cascalho ou rochas e corais (McCormick, 1995). Espécies de
Mulloidichthys forrageiam principalmente sobre substratos não consolidados e geralmente
ocorrem em regiões com depósito de sedimento ao redor de recifes (Gosline, 1984;
McCormick, 1995). Já as espécies de Parupeneus e Pseudupeneus são mais versáteis e
forrageiam também sobre substratos consolidados, ocorrendo freqüentemente sobre os
recifes (Gosline, 1984; McCormick, 1995).
Apesar de todas as espécies de Mullidae forragearem no fundo, com auxílio dos
barbilhões, as diferentes espécies podem apresentar repertórios distintos de táticas
alimentares. Há espécies que usam cinco táticas alimentares, as mais comuns sendo enterrar
o focinho no substrato ou abocar a camada superior do substrato (e.g. Lukoschek &
McCormick, 2001). Cada tática alimentar tem impacto distinto no substrato e sua fauna
associada (McCormick, 1995; Lukoschek & McCormick, 2001).
Espécies de Mullidae podem ter atividade alimentar exclusivamente diurna,
exclusivamente noturna, ou diurna e noturna (Randall, 1967; Hobson, 1974; Gosline,
1984). Quando não estão em atividade alimentar, os mulídeos formam grupos pequenos ou
médios (de 4 a 30 indivíduos) e permanecem estacionários na coluna d’água. Os grupos
inativos podem ocorrer tanto em grutas e reentrâncias no recife como em águas
relativamente abertas (Randall, 1967; Munro, 1976; Gosline, 1984).
6
Área de estudo
O Arquipélago de Fernando de Noronha se situa a cerca de 400km da costa nordeste
do Brasil. O arquipélago é formado por 21 ilhas e ilhotas, sendo que a maior delas
(Fernando de Noronha) tem cerca de 10km de comprimento e 27km2 (Figura 2) (Linsker,
2003; www.noronha.pe.gov.br). O arquipélago tem origem vulcânica e está situado sobre o
pico de uma montanha submersa cuja base está à cerca de 4.500m de profundidade. Assim,
a costa do arquipélago apresenta declive acentuado e a poucos metros da costa a
profundidade pode ultrapassar os 100m (Linsker, 2003).
Os recifes de Fernando de Noronha são compostos principalmente por rochas
cobertas por algas pardas e sedimento. A costa noroeste da ilha, denominada “Mar de
Dentro” apresenta mar calmo durante os meses de maio a novembro e é onde foi feita a
maioria das observações do presente estudo. Na costa sudeste, conhecida como “Mar de
Fora”, o mar é batido durante a maior parte do ano, ficando relativamente calmo durante os
meses de novembro a abril (Linsker, 2003).
A principal área de estudo está localizada na Praia da Conceição, no “Mar de Dentro”
(Figura 2b, 3). Esta área é composta por recife rochoso e planície arenosa adjacente. Nas
áreas de costão rochoso predominam quatro tipos de substrato: 1) algas pardas (Dictyota
sp., Dictyopteris sp. e Sargassum sp.; 2) misto de algas e areia (composto principalmente
por areia, algas coralináceas, algas verdes calcificadas e Dictyota sp.); 3) arenoso e 4)
rochoso e coralíneo (Figura 4). A profundidade varia de um a 10m, a temperatura é
constante ao longo do ano, variando ente 27-28oC e a visibilidade subaquática pode variar
entre 5-25m.
7
Figura 2. Localização do Arquipélago de Fernando de Noronha (03o54'S, 32o25'W) no
Atlântico Ocidental (a) e mapa do arquipélago (b). A principal área de estudo, Praia da
Conceição, está marcada com “*” no mapa do arquipélago.
Figura 3. Costão rochoso da Praia da Conceição, Fernando de Noronha, principal área de
estudo.
a b
8
Figura 4. Os quatro tipos de substratos estudados na Paia da Conceição: misto (a), algas
pardas (b), arenoso (c) e rochoso (d).
Foram também feitas observações e registros em vídeo e fotográficos em outras
regiões do arquipélago, como Porto de Santo Antônio, Praia do Sancho (no “Mar de
Dentro”) e Baía do Sueste (no “Mar de Fora”). As observações no Porto de Santo Antônio
foram feitas em um naufrágio a cerca de 6-8m de profundidade. Na Praia do Sancho, as
observações se concentraram na costa sul, entre 5-12m de profundidade e, na Baía do
Sueste, na costa norte entre um e dois metros de profundidade. Em todas as áreas de estudo
o fundo era semelhante, principalmente composto por algas pardas e fundo arenoso
adjacente ao recife.
b a
c d
9
Espécies estudadas
Em Fernando de Noronha (Pernambuco), ocorrem duas espécies sintópicas de
Mullidae, Mulloidichthys martinicus e Pseudupeneus maculatus (Figura 5). Mulloidichthys
martinicus ocorre no Atlântico Ocidental, da Flórida ao Rio de Janeiro, apresenta atividade
alimentar diurna e noturna e pode forragear solitária ou em pequenos a grandes grupos
(Randall, 1967; Munro, 1976; Aronson & Sanderson, 1987; Carvalho-Filho, 1999).
Pseudupeneus maculatus ocorre no Atlântico Ocidental, de Nova Jersey a Santa Catarina,
apresenta atividade alimentar diurna e forrageia solitária ou em pequenos grupos (Starck &
Davis, 1966; Munro, 1976; Carvalho-Filho, 1999). Ambas são espécies recifais,
freqüentemente encontradas sobre fundo de areia e cascalho (Starck & Davis, 1966; Munro,
1976; Carvalho-Filho, 1999).
Figura 5. Mulloidichthys martinicus (a) e Pseudupeneus maculatus (b) têm ampla
distribuição no Atlântico Ocidental, sendo sintópicos em diversos locais de Fernando de
Noronha.
Não há estudos sobre seleção de substrato para forrageio pelas duas espécies, porém,
P. maculatus parece forragear exclusivamente em substrato arenoso no Caribe (Itzkowitz,
1977; Aronson & Sanderson, 1987; J.P. Krajewski & R.M. Bonaldo obs. pess.).
b a
10
Mulloidichthys martinicus e P. maculatus alimentam-se de presas semelhantes (e.g.,
poliquetas, caranguejos, camarões, moluscos bivalves), mas apresentam diferenças nas
proporções volumétricas de itens alimentares consumidos, quando em simpatria (Randall,
1967). Ainda, M. martinicus alimenta-se de maior variedade de itens sendo,
plausivelmente, menos seletivo na escolha de itens alimentares que P. maculatus (Randall,
1967; Munro, 1976). Há um único registro de zooplâncton na dieta de M martinicus no
Caribe (Sierra et al., 1994), porém sem maiores comentários.
11
OBJETIVOS
O presente estudo examina comparativamente a atividade de forrageamento e
comportamento alimentar de M. martinicus e P. maculatus, espécies sintópicas em
Fernando de Noronha. No primeiro capítulo, são examinadas as proporções de investidas
em diferentes tipos de substratos, a freqüência alimentar, a duração das investidas no
substrato e as táticas alimentares cada espécie. No segundo capítulo, são apresentados os
primeiros registros de planctofagia para P. maculatus e um dos únicos registros de
planctofagia em Mullidae, feitos em Fernando de Noronha (Pernambuco) e na Reserva
Biológica Marinha do Arvoredo (Santa Catarina). No terceiro capítulo, é apresentado um
possível exemplo de mimetismo de proteção entre M. martinicus e Haemulon
chrysargyreum (Haemulidae) em Fernando de Noronha.
12
REFRÊNCIAS
Aronson, R. B. &. Sanderson S. L (1987). Benefits of heterospecific foraging by the
Caribbean wrasse, Halichoeres garnoti (Pisces: Labridae). Environmental Biology of
Fishhes 18, 303-308.
Barnes, R.S.K., Calow, P., Olive, P.J.W. & Golding D. W. (1993). The Invertebrates: A
New Synthesis 2nd edition. Blackwell Science Inc.
Birkeland, C. & Neudecker S. (1981). Foraging behavior of two Caribbean chaetodontids:
Chaetodon capistratus and Chaetodon aculeatus. Copeia 1981, 169-178.
Carvalho-Filho, A. (1999). Peixes: Costa Brasileira. 3a edição. Editora Melro, São Paulo.
283pp.
Froese, R. & Pauly D. (Eds.) (2005). FishBase.World Wide Web electronic publication.
www.fishbase.org.
Gosline, W. A. (1984). Structure, function, and ecology in the goatfishes (family Mullidae).
Pacific Science 38, 312-323.
Helfman, G.S., B.B. Collete & Facey D.E. (1997). The Diversity of Fishes. Blackwell
Science, Malden, 528pp.
Hobson, E.S. (1965). Diurnal-nocturnal activity of some inshore fishes in the Gulf of
California. Copeia 1965, 291-302.
Hobson, E.S., (1974). - Feeding relationships of teleostean fishes on coral reefs in Kona,
Hawaii. Fisheries Bulletin 72(4), 915-1031.
Itzkowitz, M. (1977). Social dynamics of mixed-species groups of Jamaican reef fishes.
Behavioural Ecology and Sociobiology 2, 361-384.
Krebs, C.H. (1999). Ecological methodology. 2nd edition. Addison-Welsey Educational
Publishers, Inc., Menlo Park, CA. 620pp.
13
Lehner, P.N. (1979). Handbook of ethological methods. Garland STPM Press, New York,
403pp.
Linsker, R. (ed). (2003). Arquipélago Fernando de Noronha: o paraíso do vulcão. Terra
virgem, São Paulo. 167 p.
Lukoschek, V. & McCormick, M. I. (2001). Ontogeny of diet changes in a tropical benthic
carnivorous fish, Parupeneus barberinus (Mullidae): Relationship between foraging
behaviour, habitat use, jaw size, and prey selection. Marine Biology 128, 1099-1113.
McAfee, S.T. & Morgan S.G. (1996). Resource use by five sympatric parrotfishes in the
San Blas Archipelago, Panama. Marine Biology 125, 427-437.
McCormick, M. I. (1995). Fish feeding on mobile benthic invertebrates: influence of
variability in habitat associations. Marine Biology 121, 627-637.
Munro, J.L. (1976). Aspects of the biology and ecology of Caribbean reef fishes: Mullidae
(goat fishes). Journal of Fish Biology 9, 79-97.
Randall, J. E. (1967). Food habits of reef fishes of the West Indies. Studies in Tropical
Oceanography Miami 5, 665-847.
Sazima, I. (1986). Similarities in feeding behavior between some marine and freshwater
fishes in two tropical communities. Journal of Fish Biology 29, 53-65.
Sierra, L. M., Claro, R. & Popova, O. A. (1994). Alimentación y relaciones tróficas. In
Ecología de los peces marinos de Cuba (Claro, R., ed.), pp. 263-320. Quintana Roo,
México: Centro de Investigaciones de Quintana Roo.
Starck, A.S. & Davis, W.P. (1966). Night habits of fishes of Alligator Reef, Florida.
Ichthyologica 38, 315-356.
www.noronha.pe.gov.br “Site” Oficial da Administração do Distrito Estadual de Fernando
de Noronha 2005.
14
Capítulo 1
Foraging activity and behaviour of two syntopic goatfish species
(Perciformes: Mullidae) at Fernando de Noronha Archipelago, tropical
West Atlantic
João Paulo Krajewski1*, Roberta Martini Bonaldo1, Cristina Sazima1,2 & Ivan Sazima1
1) Departamento de Zoologia e Museu de História Natural, CP 6109, Universidade
Estadual de Campinas, 13083-970 Campinas, São Paulo, Brazil
*Corresponding author. Tel: + 55-19-3255 2038; fax: +55-19-3254 2177; e-mail:
2) Departamento de Zoologia, Caixa Postal 199, Universidade Estadual Paulista, 13506-
900 Rio Claro, SP, Brazil
15
Abstract
The goatfishes (Mullidae) include about 50 specialized bottom forager fish species. The
foraging activity of the yellow goatfish, Mulloidichthys martinicus, and the spotted
goatfish, Pseudupeneus maculatus, was studied comparatively at Fernando de Noronha
Archipelago, off northeast Brazil, tropical West Atlantic, where they are syntopic. Pseudupeneus
maculatus fed over a larger variety of substrate types, had lower feeding rate, roamed more
per given time, spent less time in a feeding event, and displayed a more diverse repertoire
of feeding modes than M. martinicus. The differences in the foraging activity and
behaviour between the two species possibly minimize potential resource overlap between
them, as recorded for other syntopic mullids. Pseudupeneus maculatus had lower feeding
rate probably because it fed on larger items. Additionally, it roamed over greater distance
per time possibly because it foraged over a greater variety of substrate distributed over a
larger area than that used by M. martinicus. Notwithstanding the overall similarity between
the mullids in general, they do differ in their substrate preferences and foraging activity,
which indicates that these fishes should not be simply considered generalized soft bottom
foragers.
Key words: Mulloidichthys martinicus, Pseudupeneus maculatus, foraging behaviour,
oceanic island.
16
Introduction
Goatfishes (Perciformes: Mullidae) include about 50 species distributed in tropical and
subtropical seas. All goatfishes are zoobenthivores and forage mainly over soft sediments
(sand and mud) around reefs, oriented mostly by their tactile and chemosensorial barbels
(Gosline, 1984; McCormick, 1993, 1995; Platell et al., 1998). During their foraging, the
goatfishes disturb the substrate and, thus, change the bottom topography and the
distribution of invertebrates associated to soft sediments (McCormick, 1995). Additionally,
as a consequence of their foraging mode, the goatfishes act as nuclear species and are
frequently followed by other carnivore reef fishes that catch small preys flushed during
their substrate disturbance (Gosline, 1984; Aronson & Sanderson, 1987).
The goatfishes are common and abundant in tropical and sub-tropical shallow reefs
(Platell et al., 1998) and two or more goatfish species frequently occur in the same reef
(Gosline, 1984; Golani, 1994; McCormick, 1995). Despite their overall similarity, co-
occurring goatfish species usually differ in their foraging substrate and feeding modes
(Gosline, 1984; Golani, 1994; McCormick, 1995; Platell et al., 1998). Such differences are
believed to minimize food overlap between co-occurring goatfishes (Labropoulou &
Eleftheriou, 1997).
Two species of goatfishes, the yellow goatfish, Mulloidichthys martinicus (Cuvier), and
the spotted goatfish, Pseudupeneus maculatus (Bloch) are syntopic at Fernando de Noronha
Archipelago, off northeast Brazil, tropical West Atlantic. The yellow goatfish is a common reef
species in the Western Atlantic and ranges from Florida to South-eastern Brazil (Carvalho-
Filho, 1999; Humann & DeLoach, 2002). It feeds both at daytime and night (Randall, 1967;
Munro, 1976; Aronson & Sanderson, 1987) and forages either solitarily or in groups
(Carvalho-Filho, 1999). The spotted goatfish also is a common reef fish species in the
17
Western Atlantic and ranges from New Jersey to South Brazil (Carvalho-Filho, 1999;
Humann & DeLoach, 2002). It feeds only during the day, either solitary or in small groups
(Starck & Davis, 1966; Munro, 1976).
Mulloidichthys martinicus and P. maculatus feed on similar items, but P. maculatus
generally has, in its stomach, a higher volumetric content of mobile preys, such as crabs
and shrimps, and also feeds on small fishes (Randal, 1967). The relative high proportion of
mobile and large prey in the stomach content of the spotted goatfish is possibly related to
the fact that all species within the genus Pseudupeneus have strong jaw teeth, as well as a
relatively long snout, both features apparently suited to catch large and active prey
(Gosline, 1984).
The foraging activity of M. martinicus and P. maculatus was studied comparatively at
Fernando de Noronha Archipelago. The present study addressed the five following
questions: 1) Do the two species use the same foraging substrate? 2) Do their feeding rates
differ? 3) Are there differences in the distance travelled per given period while foraging? 4) Does
the time spent in a feeding event differ between the two species? 5) Do their feeding modes differ?
Materials and methods
The study was conducted at the Fernando de Noronha Archipelago (03o50’ S; 32o25’
W), 345km off north-eastern Brazil, from June to July 2002 and June 2003. The main study
site was an area of about 200 X 30 m at the Praia da Conceição, on the west side of
Fernando de Noronha main island. The bottom was composed of rocky reefs covered
mostly by brown algae (Dictyota sp., Dictyopteris sp., and Sargassum sp.), mixed substrate
(composed mostly with sand and an assemblage of corallinaceous and green calcified algae,
as well as Dictyota sp) and adjacent sandy areas. During the study the depth ranged 1-12 m,
visibility ranged 8-30 m and water temperature was 27-28oC.
18
The foraging activity of the two goatfishes was observed over 55 non-consecutive days
while snorkelling and scuba-diving. During observational sessions of 60-150 min, “focal
animal” and “all occurrences” samplings (Lehner, 1979) were used in a total of 640 min of
direct observation. All observations were conducted at daytime from 0930h to 1730h.
Selection of foraging substrate and feeding rates of the two species were assessed by
following individuals for periods of 3 to 5 min and counting the number of feeding events
on four substrate types: 1) brown algae; 2) mixed (algae and sand); 3) sandy, and 4) hard
substrate (rocks and stony corals). Each feeding event started when the goatfish’s mouth or
snout touched the substrate and ended when it left the substrate. The distance travelled per
given time was assessed by following foraging individuals for 3 to 5 min and measuring the
distance the fish moved within this time. Individuals were not followed over successive
periods to avoid the risk of biased samples (Birkeland & Neudecker, 1981). Data used for
substrate selection comparisons was collected in June 2003 and at the same study site, and
therefore the substrate availability was the same for all individuals studied. Data on feeding
rates and distance travelled per time was collected in June and July 2002 and June 2003 at
the same study site, using similar observational times for the two species in the consecutive
years.
The relative abundance of the four foraging substrate types was assessed by video
recording eight 30 x 1.5 m transects (modified from Birkeland & Neudecker, 1981).
Transects started on the rocky shore and ended on the sandy area where the two goatfishes
were still recorded feeding. Still frames were taken from the video recordings and the
availability of the four substrates types considered was thus estimated (brown algae, sand,
mixed and hard substrate).
About two hours of foraging activity of the two goatfishes were video-recorded and
photographed at four different sites in the archipelago. In all sites the bottom was similar to
19
that of Praia da Conceição. A total of 128 feeding events of 19 individuals of M. martinicus
and 118 feeding events of 21 individuals of P. maculatus were analyzed. Time spent during
each feeding event was estimated by analyzing the video records frame by frame. Time
average spent in each feeding event was calculated for each individual recorded, and the
total mean between all individuals was calculated for each species. The feeding modes of
the two goatfish species were quantified analyzing the video records and photographs, and
a name for each mode is here proposed.
Results
Mulloidichthys martinicus and P. maculatus differed in all studied features (Figures1
and 2 and Table I). Mulloidichthys martinicus fed over sandy and mixed (algae and sand)
substrate, but preferred sandy bottom, whereas P. maculatus fed over sandy, mixed and
brown algae substrate, but preferred mixed substrate (Figure 1). Both species were also
recorded to feed on rubble substrate (composed of pieces of dead calcified algae of about 2-
20 mm, and sand) at study sites other than the main site at the Praia da Conceição.
20
Figure 1. Frequency of feeding events of Mulloidichthys martinicus and Pseudupeneus
maculatus on each substrate type, and the actual relative abundances of those substrates at
Fernando de Noronha Archipelago.
Mulloidichthys martinicus had higher feeding rate, roamed less per given time, and spent
more time during each feeding event than P. maculatus (Table I).
21
Table I. Comparisons between Mulloidichthys martinicus and Pseudupeneus maculatus at
Fernando de Noronha Archipelago: Student t test for feeding rates and distance roamed per
time and Mann-Whitney for time spent during each feeding event. The parentheses indicate
the total number of bouts for feeding rate and distance roamed per time data and number of
individuals for time spent in each feeding event data.
M. martinicus P. maculatus p
Feeding rate
(feeding events/min)
3.42
(n=30)
1.99
(n=53)
<0.0001
Distance roamed per time
(m/min)
2.99
(n=26)
5.39
(n=33)
<0.0001
Time spent in each feeding event
(s)
1.56
(n=17)
0.61
(n=16)
<0.0001
The feeding modes displayed by the two species were here classified into six categories
(Figure 2) (modified from McCormick, 1995): 1) excavate: burrowing with snout, the body
axis usually more than 30o relative to substrate; 2) push: moving the snout against the
substrate dislodging it horizontally, the body axis usually less than 20o relative to substrate;
3) drift: barbels actively pushing the top layer of the sediment; 4) bite: capture of prey using
only jaws’ movement; 5) pursuit: short distance chase (about 5 -30cm) of a mobile prey
flushed during the substrate disturbance; 6) ambush: striking a prey after remaining
stationary for 1 to 5 sec camouflaged on the substrate.
Pseudupeneus maculatus displayed the six feeding modes, whereas M. martinicus
displayed four of them: excavate, push, drift and bite. The two species additionally differed
in the frequency that they displayed each feeding mode, since M. martinicus mostly pushed
and excavated, whereas P. maculatus mostly bit and drifted (Figure 3).
22
Figure 2. A hypothetic goatfish displaying the six feeding modes (from the left to the right)
recorded for Mulloidichthys martinicus and Pseudupeneus maculatus at Fernando de
Noronha Archipelago. The former species displayed only excavate, push, drift and bite,
whereas the latter one displayed all modes. Based on video frames and photographs.
23
Figure 3. Frequency of each feeding mode by Mulloidichthys martinicus and
Pseudupeneus maculatus at Fernando de Noronha Archipelago.
Discussion
All goatfish exploit the bottom, especially soft substrates, using their barbels and snout
(Gosline, 1984; McCormick, 1995). Thus, sympatric species have the potential to overlap
in food resources use. However, in spite of their overall similarity of feeding habits,
goatfish species usually differ in the substrate they forage and in their feeding modes
(Gosline, 1984; Golani, 1994; McCormick, 1995; Platell et al., 1998). In the present study,
M. martinicus and P. maculatus differed in all analyzed features. In the Caribbean, these
two goatfish species are known to occur at different depths (Munro, 1976). The differences
in foraging activity and depth distribution of both species may minimize the food resource
overlap between them where they live in sympatry. Additionally, M. martinicus and P.
maculatus differ in their activity period, since the former is known to feed at night as well
(Randall, 1967; Munro, 1976). However, differences in activity period may not be very
important to avoid food resource overlap because the consumption of one food resource at
night at a given site reduces its availability during the day at this site (Schoener, 1974).
24
Although goatfish species usually differ in their feeding activity, which is regarded as a
mechanism of avoid resource overlap and competition, data from the present study do not
necessarily points that differences in foraging behaviour of M. martinicus and P. maculatus
is a mechanism to minimize competition (Connel, 1980). Differences between these two
species may also have resulted from past competition or simply by differences arisen
during their speciation (Connel, 1980).
Even though there is no study that examines foraging substrate use by M. martinicus or
P. maculatus, in the Caribbean these two species seem to feed almost exclusively over
sandy bottoms (Itzkowitz, 1977; Aronson & Sanderson, 1987; Krajewski & Bonaldo pers.
obs.). The present study findings for M. martinicus are similar to those obtained in the
Caribbean (Aronson & Sanderson, 1987), since it foraged only over soft substrates (sandy,
mixed, and rubble) mainly composed of sand. On the other hand, in this study P. maculatus
foraged over a wider range of substrate types than in the Caribbean (Itzkowitz, 1977;
Aronson and Sanderson, 1987; Krajewski and Bonaldo, pers. obs.), especially in Jamaica,
where this goatfish was recorded to forage exclusively on sandy substrate (Itzkowitz,
1977). Additionally, part of the brown algae substrate in our main study site is located in
very shallow waters (about 5 to 40 cm deep) and may be inaccessible for P. maculatus
during low tides and under the impact of waves. Thus, the preference for brown algae
substrate here recorded for P. maculatus may be underestimated and this substrate may be
more used for foraging by this goatfish than observed in the present study. These results
indicate that P. maculatus and species of the closely related genus Parupeneus (Gosline,
1984; McCormick, 1995) have the potential to forage mostly in reef areas and are not sandy
bottom specialists. Perhaps P. maculatus is known as foraging almost exclusively on soft
substrates in some areas of the Caribbean because many coral reefs are present there, and
algal cover is not as abundant as it is at Fernando de Noronha (J. P. Krajewski and R. M.
25
Bonaldo per. obs.). Similar questions related to bottom differences in the Caribbean and on
Brazil’s coast were raised about the cleaning role of the french angelfish, Pomacanthus
paru (Bloch), in these two regions (Sazima et al., 1999).
Species of the genera Pseudupeneus and Parupeneus are believed to feed on larger and
more mobile preys when compared with other goatfish species (Gosline, 1984).
Pseudupeneus maculatus is known to feed on more mobile and larger preys than M.
martinicus (Randal, 1967). In the present study, P. maculatus was occasionally observed
mouthing and ingesting large preys (up to 4 cm total length) such as fishes and crustaceans.
Thus, P. maculatus probably had lower feeding rate when compared with M. martinicus
due to its feeding on large food items in a higher proportion. The present study indicates
that the yellow goatfish feeds continuously on small preys, whereas the spotted goatfish
feeds on larger preys, thus in lesser amount.
Pseudupeneus maculatus foraged over three substrates types that altogether span about
93% of the total area of the study site, whereas the substrates used by M martinicus span
about 47% of the same site (Fig. 1). The yellow goatfish probably roamed for lesser
distances because the substrate it used was concentrated on the sand flat.
The lesser time spent in each feeding event by P. maculatus possibly is related to the
ability of species of the genera Pseudupeneus and Parupeneus to catch large and mobile
preys using their strong teeth and long snout. These may help goatfishes of the two genera
immobilise and ingest preys quickly and, thus, they do not spend time immobilising preys
pushing them against the substratum as M. martinicus have to do. This suggestion seems
supported by the feeding modes displayed by the two goatfishes here studied. Whereas M.
martinicus fed frequently burrowing its snout into the soft bottom, P. maculatus displayed a
variable feeding repertoire and frequently caught either immobile or mobile preys using
quick bites only.
26
In spite of their overall similarity in feeding behaviour, each goatfish species seem to
have its particular way to explore the bottom. Some species are most often associated to
soft bottoms and usually use their snout and barbels to excavate and search for buried
preys, whereas other species forage on a wide range of bottom types and display a variable
feeding repertoire (Gosline, 1984; McCormick, 1995). In the present study, M. martinicus
was mostly associated with soft bottoms and displayed feeding modes similar to that
commonly used by other soft-sediment specialists, such as several Mulloidichthys species
from the Pacific (Gosline, 1984; McCormick, 1995; Randall et al., 1997). On the other
hand, P. maculatus displayed a variable feeding repertoire, similarly to other reef species of
the genera Pseudupeneus and Parupeneus (Gosline, 1984; McCormick, 1995).
Additionally, the spotted goatfish displayed two feeding modes (pursuit and ambush) which
are not recorded for any other goatfish species (Gosline, 1984; McCormick, 1995). While
ambushing, the colour of the spotted goatfish matched that of the substrate it hunted over,
working as a camouflage to disguise the predator. The colours most often displayed by P.
maculatus (light brown with yellow and dark brown spots) are similar to those of the brown
algae substrate, abundant in all study sites, which may facilitate this goatfish’s camouflage
during ambush and increase its hunting success on visually guided and quickly fleeing
prey.
Differences in the frequency of feeding modes between the two species indicate that
each of them has a different impact and disturbance on the substrate and the hidden prey.
Mulloidichthys martinicus caused greater disturbance in soft sediments than P. maculatus,
which, additionally, cause even greater disturbance while foraging in large groups.
Overall, Pseudupeneus maculatus may be characterized as a versatile bottom predator
that forages both on soft and hard substrates. Additionally, the spotted goatfish is known to
feed on zooplankton while foraging in groups (Krajewski & Bonaldo, submited), which
27
supports its characterization as a versatile predator. On the other hand, M. martinicus seems
to be a soft bottom specialist, even if recorded to feed on zooplankton in the Caribbean
(Sierra et al., 1994).
The present study and other recent findings indicate that goatfishes can not be simply
characterized as soft bottoms specialists (Gosline, 1984, McCormick, 1995). Different
species of goatfishes may have different impacts on soft, algae and hard bottom types and
the associated organisms, either by differential consumption or disturbance. Additionally,
goatfishes are ubiquitous nuclear species, and are followed by a number of generalist
predators (Aronson & Sanderson, 1987, Soares & Barreiros, 2003, Sazima et al., 2005).
Thus, differences in the foraging behaviour and activity of goatfishes may also lead to
differences in the species that associate with each goatfish species, and in the possible
benefits gained by the follower species.
Acknowledgements
We thank JM Silva-Jr (Centro Golfinho Rotador) and the Projeto Tamar (C Bellini and A
Grossman) for logistical support at Fernando de Noronha Archipelago; the Ibama (MA
Silva) for the always warm reception, logistical support and issuing study permits at
Fernando de Noronha Archipelago; the Águas Claras, Atlantis, and Noronha Divers diving
centres for allowing free use of their facilities; S. R. Floeter, F. P. P. Leite and V. S. Uieda
for suggestions on the manuscript; Guimarães-Jr. for help with the statistical analysis; the
CAPES, CNPq, FAEP-Unicamp, FAPESP, and PROAP-UNICAMP for essential financial
support.
28
References
Aronson, R. B. & Sanderson, S. L. (1987). Benefits of heterospecific foraging by the
Caribbean wrasse, Halichoeres garnoti (Pisces: Labridae). Environmental Biology of
Fishes 18, 303-308.
Birkeland, C. & Neudecker, S. (1981). Foraging behavior of two Caribbean chaetodontids:
Chaetodon capistratus and C. aculeatus. Copeia 1981, 169-178.
Carvalho-Filho, A. (1999). Peixes da Costa Brasileira, 3a ed. São Paulo: Editora Melro.
Connell, J. H. (1980). Diversity and coevolution of competitors, or the ghost of competition
past. Oikos 35, 131-138.
Golani, D. (1994). Niche separation between colonizing and indigenous goatfish (Mullidae)
along the Mediterranean coast of Israel. Journal of Fish Biology 45, 503-513
Gosline, W. A. (1984). Structure, function, and ecology in the goatfishes (family Mullidae).
Pacific Science 38, 312-323.
Humann, P. & Deloach, N. (2002). Reef fish identification: Florida, Caribbean, Bahamas.
Jacksonville, Florida: New World Publications.
Itzkowitz, M. (1977). Social dynamics of mixed-species groups of Jamaican reef fishes.
Behavioural Ecology and Sociobiology 2, 361-384.
Labropoulou, M. & Eleftheriou, A. (1997). The foraging ecology of two pairs of congeneric
demersal fish species: importance of morphological characteristics in prey selection.
Journal of Fish Biology 50, 324-340.
Lehner, P.N. (1979). Handbook of ethological methods. New York: Garland STPM Press.
McCormick, M. I. (1993). Development and changes at settlement in the barbel structure of
the reef fish, Upeneus tragula (Mullidae). Environmental Biology of Fishes 37, 269-282.
McCormick, M. I. (1995). Fish feeding on mobile benthic invertebrates: influence of
variability in habitat associations. Marine Biology 121, 627-637.
29
Munro, J. L. (1976). Aspects of the biology and ecology of Caribbean reef fishes: Mullidae
(goat fishes). Journal of Fish Biology 9, 79-97.
Platell, M. E., Potter, I. C. & Clarke, K. R. (1998). Do the habitats, mouth morphology and
diets of the mullids Upeneichthys stotti and U. lineatus in coastal waters of south-
western Australia differ? Journal of Fish Biology 52, 398-418.
Randall, J. E. (1967). Food habits of reef fishes of the West Indies. Studies in Tropical
Oceanography Miami 5, 665-847.
Randall J. E., Allen, G. R. & Steene, R. C. (1997). Fishes of the Great Barrier Reef and
Coral Sea. Honolulu, Hawaii: University of Hawaii Press.
Sazima, C., Bonaldo, R. M., Krajewski, J. P. & Sazima, I (2005). The Noronha wrasse: a
“Jack-of-al-trades” folower. Aqua, Journal of Ichthyology and Aquatic Biology. 9(3),
97-108
Sazima, I., Moura, R. L. & Sazima C. (1999), Cleaning activity of juvenile angelfish,
Pomacanthus paru, on the reefs of the Abrolhos Archipelago, western South Atlantic.
Environmental Biology Fishes. 56, 399-407.
Schoener, T. W. (1974). Resource partitioning in ecological communities. Science 185, 27-
39.
Sierra, L. M., Claro, R. & Popova, O. A. (1994). Alimentación y relaciones tróficas. In
Ecología de los peces marinos de Cuba (Claro, R., ed.), pp. 263-320. Quintana Roo,
México: Centro de Investigaciones de Quintana Roo.
Soares, M. S. C. & Barreiros, J. P. (2003). Following associations with the striped mullet
Mullus surmuletus Linneaus, 1758 (Perciformes: Mullidae) from the Azores. Aqua,
Journal of Ichthyology and Aquatic Biology 7, 139-144.
Starck, A.S. & Davis, W.P. (1966). Night habits of fishes of Alligator Reef, Florida.
Ichthyologica 38, 315-356.
30
Capítulo 2
Plankton-picking by the goatfish Pseudupeneus maculatus (Mullidae), a
specialized bottom forager
João Paulo Krajewski1*, Roberta Martini Bonaldo1
1) Departamento de Zoologia e Museu de História Natural, CP 6109, Universidade
Estadual de Campinas, 13083-970 Campinas, São Paulo, Brazil (www.unicamp.br)
*Corresponding author. Tel: + 55-19-3255 2038; fax: +55-19-3254 2177; e-mail:
31
Abstract
The goatfishes (Mullidae) include about 50 specialized bottom forager fish species.
Juveniles and adults of the spotted goatfish Pseudupeneus maculatus were recorded feeding
on plankton on rocky reefs off Southern and Northeastern Brazil, respectively.
Key words: benthic foraging, plankton feeding, zooplankton, opportunistic behavior, Western
Atlantic
Plankton-feeding is a widespread foraging mode among reef fishes (Hobson, 1991).
Some species feed on plankton only during their larval and early juvenile stages, while
others remain obligate planktivores when adults (Hobson, 1991). Fishes specialized for
picking zooplankton generally have protrusible jaws and short bodies, and typically form
aggregations in the water column in areas where zooplankton is abundant (Hobson &
Chess, 1978, 1986; Hobson, 1991). Some bottom-foraging reef fish are reported to feed on
plankton or particulate food occasionally and under particular circumstances (e.g. Glasby &
Kingsford, 1994; Sazima & Sazima, 2001; Sazima et al., 2003).
The goatfishes (Mullidae) include about 50 specialized bottom-foraging fishes that use
their well developed chemo-sensorial barbels to probe and disturb the substrate while
feeding (Munro, 1976; McCormick, 1995). Differently from several other reef fish families
that have strong benthic association (e.g., damselfishes, wrasses, snappers, and groupers)
and include both bottom and plankton-foraging species, no goatfish is known as primarily
planktivorous when adult (McCormick, 1995).
The goatfishes are reported as plankton-eaters only during their larval stage, and after a
drastic and quick change during their settlement they develop a set of features that seems to
32
constrain them to a benthic foraging habit (Gosline, 1984; McCormick, 1993, 1995).
However, there is a report on zooplankton in the diet of the yellow goatfish, Mulloidichthys
martinicus (Cuvier) from Cuba (Sierra et al., 1994), and a record of foraging in the water
column by the goatfish Parupenes barberinus (Lacepède) in Australia (Lukoschek &
McCormick, 2001), both records without further comments.
Groups of the spotted goatfish, Pseudupeneus maculatus (Bloch), were recorded feeding
on plankton on the rocky reefs of the Marine Biological Reserve of Arvoredo (about
27o17’S, 48o18W), on the coast of Santa Catarina, southern Brazil, and at Fernando de
Noronha Archipelago (03o50’S, 32o25’W), about 345 km off northeast Brazil. The
observations at the Marine Biological Reserve of Arvoredo were conducted in February
2004, the depth at the study site was 5-15 m and horizontal visibility was 4-10 m. The
plankton-feeding behaviour of groups of juvenile spotted goatfish was observed and
photographed during eight scuba-diving sessions totalling eight hours. The observations at
Fernando de Noronha Archipelago were conducted in June 2003 and the depth was 15 m
and horizontal visibility 30 m. A mixed group of P. maculatus and the damselfish
Abudefduf saxatilis (Linnaeus) was video recorded while they foraged on suspended
particles and zooplankton.
At the Marine Biological Reserve of Arvoredo a total of 13 groups of juvenile spotted
goatfish were recorded. Each group was composed of 12-16 individuals of 5-10 cm total
length (LT), most larger than 7 cm LT. All groups were constantly feeding on the bottom
over brown algae substrate and the plankton-picking foraging was recorded when the
individuals suddenly rose up to one m from the bottom and started to pick small
crustaceans in the water column (Fig. 1). The fishes spent about 4-10 s feeding on plankton
before they resumed their habitual bottom-feeding. During the plankton-feeding the fish
protruded their jaws with their barbels retracted and swam back and forth to pick off
33
individual plankters. In all instances of plankton-feeding a cloud of small crustaceans was
visible. No adult individuals of the spotted goatfish were recorded feeding on plankton at
the Marine Biological Reserve of Arvoredo, most were foraging solitarily on the bottom.
Figure 1. A group of juvenile, recently settled goatfish, Pseudupeneus maculatus, rising in
the water column to feed on plankton. The white spots in the water column are due to the
flash light reflected on the zooplankton.
At Fernando de Noronha Archipelago a mixed group of about 10 P. maculatus of 15-20
cm LT and 33 A. saxatilis, was recorded. The group was feeding on the bottom and
plankton-feeding was recorded when the goatfish started to pick small crustaceans about 2-
5 cm above a brown algae agglomerate that was floating about 20-30 cm above the sandy
bottom (Fig. 2). The plankton-picking lasted about 15 s and was displayed by six out of the
10 individuals. During plankton-feeding the fishes protruded their jaws with their barbels
retracted and swam back and forth picking off individual plankters and other particles in the
water column. A copy of the video records is deposited as a voucher in the Museu de
História Natural of the Universidade Estadual de Campinas (ZUEC tape #21)
34
Figure 2. A mixed group of the goatfish Pseudupeneus maculatus and the damselfish
Abudefduf saxatilis. The goatfish are feeding on drifting crustaceans, close to the bottom.
Note the protruded mouth of the individual on the left side of the picture (taken from a
video frame).
When goatfish settle, they go through a drastic and quick morphological and colour
change. Pelagic goatfish are usually silvery and change to adult colour shortly after
settlement. The chemosensorial barbels also change drastically during settlement, and they
are recorded to increase up to 52% in length in a 12h-period (McCormick, 1993). The
barbels are not used during the larval phase, but are frequently used to detect and
manipulate prey once the fish is settled (McCormick, 1993). Also, goatfishes are recorded
to change behaviour from schooling with other fish species under drifting debris before and
during settlement, to foraging in small groups or solitarily on the bottom after settlement
(MacCormick, 1993). For the spotted goatfish, the metamorphosis from pelagic larva to
benthic adolescence occurs when the fish is about 41mm standard length (LS), but may be
35
delayed until 61 mm LS (Munro, 1976). All individuals observed foraging on plankton at
the Marine Biological Reserve of Arvoredo were certainly post settlement, since they were
constantly foraging on the bottom and had chemosensorial barbels and colour similar to
that of adults (Figure 1). Since all individuals observed were about 5-10 cm LT, they must
be considered recently settled and this proximity to their larval phase may have favoured
the occurrence of plankton-feeding behaviour, as these fish displayed this behaviour in the
larval stage not long before the phase recorded here. The much higher frequency of
foraging in the water column by small individuals (< 12 cm LT) of P. barberinus when
compared with medium and large (> 12 cm LT) ones (Lukoschek & McCormick, 2001)
lends support to this latter suggestion.
The spotted goatfish seldom swims in open water, and even when it spawns, it generally
rises a shorter distance more quickly in the water column than other reef fishes, a strategy
believed to be a predator avoidance behaviour (Colin & Clavijo, 1978). The closeness of
the drifting crustaceans and the bottom, in both our records, may have favoured the
plankton-feeding behaviour, since the spotted goatfish could feed on the plankton without
having to rise too far from the substrate and thus be exposed to predators. Additionally, as
the plankton was close to the substrate, the goatfish could perceive it as an alternative and
rich food resource (Norrbin & Bamstedt, 1984) during their usual bottom-feeding.
The foraging in groups may also have contributed to the plankton-picking behaviour. In
the example where the spotted goatfish was feeding together with A. saxatilis, the fish
group seems to have chased the small crustaceans and concentrated them over the brown
algae mass. This likely helped the goatfish perceive the drifting crustaceans during their
usual bottom-feeding. Additionally, when foraging in groups the fish are safer from
predators (Brock & Riffenburgh, 1960), so they may be more prone to rise in the water
column. Even a solitary goatfish could pick off plankters where plankton is abundant and
36
close to the bottom in reef areas that provide shelter against predators (Colin & Clavijo,
1978).
The goatfishes suffer a major change during their settlement and develop a set of
features that seems to constrain them to benthic foraging (Gosline, 1984; McCormick,
1993, 1995). Even if they exhibit several features exclusive to bottom-foragers, the
goatfishes have protrusible jaws, a characteristic shared with specialized plankton-picking
fish species (Hobson, 1991). The protrusible jaws seem to be the main feature of the
spotted goatfish that enables it to feed on plankton and drifting particles, and several other
bottom-foraging species with protrusible jaws are likely candidates to feed occasionally on
plankton. The records of feeding in the water column by P. barberinus (Lukoschek &
McCormick, 2001), plankton-picking by the bottom-foraging butterflyfish Chaetodon
striatus (Linnaeus) (Sazima & Sazima, 2001) and dolphins’ offal picking by chub,
Kyphosus sectatrix (Linnaeus), and doctorfish, Acanthurus chirurgus (Bloch), (Sazima et
al., 2003) lend support to our suggestion. Small groups of the barber goby, Elacatinus
figaro (Sazima et al., 1996), were also recorded by us picking off plankton about 50 cm
above the bottom in Rio de Janeiro, Southeastern Brazil. This goby is a specialized, station-
based cleaner that may feed over the substrate (Sazima et al., 1996). There was a plankton
bloom on the occasion, which further strengthens our suggestion about particulate-feeding
by bottom-foraging fishes when plankton is abundant, even if it is an ephemeral resource
(see also Sazima & Sazima, 2001).
Additionally, facultative planktivorous fishes are recorded to increase their ingestion of
drifting particles when there is a bloom of highly nutritious particles in the water column
(MacCormick, 2003; Pratchett et al., 2001). There is evidence that supplementary food,
especially coral propagules, increases growth, improves nutritional and physiological
condition and increases reproductive success (McCormick, 2003; Pratchett et al., 2001).
37
Similarly, bottom-foraging fishes that occasionally feed on dolphins’ offal may obtain a
high amount of protein, since the dolphin offal is composed mainly of partially digested
fish and squid and live roundworms (Silva-Jr et al., 2004, 2005). This protein gain should
be especially advantageous to herbivorous fish (Wilson, 2002; Wilson et al., 2003), which
usually consume food with low protein content (Choat, 1991). This may explain why the
only bottom foragers recorded feeding on dolphin offal are herbivores (Sazima et al.,
2003). Therefore, it seems that bottom foraging individuals may change their usual feeding
behaviour to take advantage of rich and ephemeral food resources, such as lipid-rich coral
propagules (McCormick, 2003; Pratchett et al., 2001), protein-rich dolphin offal (Sazima et
al. 2003; Silva-Jr et al., 2004, 2005) and energy-rich plankton (Norrbin & Bamstedt, 1984;
this study).
The benefit of eating drifting particles by bottom foraging fishes may cause a “dilemma”
in terms of predation: risk in the water column versus feeding on a nutritious food. This
dilemma may be lessened by feeding on drifting particles only in “safe situations”, such as
when in groups and close to the bottom (Brock & Riffenburgh, 1960; Lukoschek &
McCormick, 2001).
The picking of drifting particles by bottom foragers is probably restricted to particular
circumstances that favour an abundance of such particles, such as mass spawning of reef
invertebrates (McCormick, 2003; Pratchett et al., 2001), voiding of faeces and vomit by
fishes and marine mammals (Arbuto-Oropeza, 2000; Sazima et al., 2003), disturbance of
soft substrates (Glasby & Kingsford, 1994), and occasional plankton blooms.
Acknowledgements: We thank J. M. Silva-Jr and the Centro Golfinho Rotador for logistics
and other help, S. R. Floeter, F. P. P. Leite, I. Sazima, and V. S. Uieda for suggestions on the
manuscript; M. S. Maule for helping in the field; the Atlantis diving center for allowing free use
38
of its facilities, the Ibama for issuing study permits at the Fernando de Noronha National
Marine Park; the CAPES, CNPq, FAPESP and PROAP-Unicamp for financial support.
References
Aburto-Oropeza, O., Sala, E. & Ortiz, C. (2000). Feeding behaviour, habitat use and
abundance of the angelfish Holacanthus passer in the southern sea of Cortez.
Environmental Biology of Fishes 57, 435-442.
Brock, V. E. & Riffenburgh, R. H. (1960). Fish schooling: a possible factor in reducing
predation. Journal du Conseil International Pour L’ Exploration de la Mer 25, 307-317.
Choat J. H. (1991). The biology of herbivorous fishes on coral reefs. In The ecology of
fishes on coral reefs (Sale, P. F., ed), pp. 120-155. San Diego, California: Academic
Press Inc.
Colin, P. & Clavijo, E. (1978). Mass spawning by the spotted goatfish, Pseudupeneus
maculatus (Bloch) (Pisces: Mullidae). Bulletin of Marine Science 28, 780-782.
Glasby, T. M. & Kingsford, M. J. (1994). Atypichthys strigatus (Pisces: Scorpididae): an
opportunistic planktivores that responds to benthic disturbances and cleans other fishes.
Australian Journal of Ecology 19, 385-394.
Gosline, W. A. (1984). Structure, function, and ecology in the goatfishes (family Mullidae).
Pacific Science 38, 312-323.
Hobson, E. S. (1991). Trophic relationships of fishes specialized to feed on zooplankters
above coral reefs. In The ecology of fishes on coral reefs (Sale, P. F., ed), pp. 69-93. San
Diego, California: Academic Press Inc.
Hobson, E. S. & Chess, J. R. (1978). Trophic relationships among fishes and plankton in
the lagoon at Enewetak Atoll, Marshall Islands. Fisheries Bulletin 76, 133-153.
39
Hobson, E. S. & Chess, J. R. (1986). Diel movements of resident and transient zooplankters
above lagoon reefs at Enewetak Atoll, Marshall Islands. Pacific Science 40, 7-26.
Lukoschek, V. & McCormick, M. I. (2001). Ontogeny of diet changes in a tropical benthic
carnivorous fish, Parupeneus barberinus (Mullidae): relationship between foraging
behaviour, habitat use, jaw size, and prey selection. Marine Biology 128, 1099-1113.
McCormick, M. I. (1993). Development and changes at settlement in the barbel structure of
the reef fish, Upeneus tragula (Mullidae). Environmental Biology of Fishes 37, 269-282.
McCormick, M. I. (1995). Fish feeding on mobile benthic invertebrates: influence of
variability in habitat associations. Marine Biology 121, 627-637.
McCormick, M. I. (2003). Consumption of coral propagules after mass spawning enhances
larval quality of damselfish through maternal effects. Oecologia 136, 37-45.
Munro, J.L. (1976). Aspects of the biology and ecology of Caribbean reef fishes: Mullidae
(goat fishes). Journal of Fish Biology 9, 79-97.
Norrbin, F. & Bamstedt, U. (1984). Energy contents in benthic and planktonic invertebrates
of Kosterfjorden, Sweden. A comparison of energetic strategies in marine organism
groups. Ophelia 23, 47-64.
Pratchett M. S., Gust N., Goby G., & Klanten, S. O. (2001). Consumption of coral
propagules represents a significant trophic link between corals and reef fish. Coral Reefs
20, 13-17.
Sazima, C. & Sazima, I. (2001). Plankton aggregation and occasional cleaning by adult
butterflyfish, Chaetodon striatus (Chaetodontidae), in Southwestern Atlantic. Cybium
25, 145-151.
Sazima, I., Sazima, C. & Silva-Jr, J. M. (2003). The cetacean offal connection: feces and
vomits of spinner dolphins as a food source for reef fishes. Bulletin of Marine Science
72, 151-159.
40
Sazima, I., Moura, R.L. & Rosa, R. S. (1996). Elacatinus figaro sp. n. (Perciformes:
Gobiidae), a new cleaner goby from the coast of Brazil. Aqua Journal of Ichthyology
and Aquatic Biology 2, 33-38.
Sierra, L. M., Claro, R. & Popova, O. A. (1994). Alimentación y relaciones tróficas. In
Ecología de los peces marinos de Cuba (Claro, R., ed.), pp. 263-320. Quintana Roo,
México: Centro de Investigaciones de Quintana Roo.
Silva-Jr, J. M., Pandolfo, L. J. & Sazima, I. (2004). Vomiting behavior of spinner dolphin
(Stenella longirostris) and squid meals. Aquatic Mammals. 30(2), 271-274.
Silva-Jr, J. M., Silva, F. J. L. & Sazima, I. (2005). Rest, nurture, sex, release and play:
diurnal underwater behaviour of the spinner dolphin at Fernando de Noronha
Archipelago, SW Atlantic. Aqua Journal of Ichthyology and Aquatic Biology 9(4), 161-
176.
Wilson, S. (2002). Nutritional value of detritus and algae in blenny territories on the Great
Barrier Reef. Journal of Experimental Marine Biology and Ecology 271, 155-169.
Wilson, S. K., Bellwood, D. R., Choat, J. H. & Furnas, M. J. (2003). Detritus in the
epilithic algal matrix and its use by coral reef fishes. Oceanography and Marine
Biology: an Annual Review 41, 279-309.
41
Capítulo 3
The association of the goatfish Mulloidichthys martinicus with the grunt
Haemulon chrysargyreum: an example of protective mimicry
João Paulo Krajewski1*, Roberta Martini Bonaldo1, Cristina Sazima1,2 & Ivan Sazima1
1) Departamento de Zoologia e Museu de História Natural, CP 6109, Universidade
Estadual de Campinas, 13083-970 Campinas, São Paulo, Brazil (www.unicamp.br)
*Corresponding author. Tel: + 55-19-3255 2038; fax: +55-19-3254 2177; e-mail:
2) Departamento de Zoologia, Caixa Postal 199, Universidade Estadual Paulista, 13506-
900 Rio Claro, SP, Brazil
42
Abstract
A presumed example of protective mimicry between the yellow goatfish, Mulloidichthys
martinicus (Mullidae) and the smallmouth grunt, Haemulon chrysargyreum (Haemulidae)
is described from Fernando de Noronha Archipelago, NE Brazil. The goatfish and the grunt
share a similar overall shape and colour pattern. We found that these two species regularly
form mixed schools around reefs. Additionally, when chased small groups of yellow
goatfish join schools of smallmouth grunts and behave like them. The colour and shape
resemblances between the two species enable their mixed schooling, and enhance the
protection against visually oriented predators for both of them. Thus, we suggest that the
protective association herein reported for the goatfish and the grunt may be considered as a
“social mimicry”, since neither species is venomous, poisonous or strongly armed.
Furthermore, we suggest that additional instances of social mimicry may involve the yellow
goatfish and other striped Haemulon species.
Key words: Protective mimicry, social mimicry, mixed schooling, Mulloidichthys
martinicus, Haemulon chrysargyreum.
43
Resumo
Descrevemos aqui um possível exemplo de mimetismo de proteção entre o saramunete
Mulloidichthys martinicus (Mullidae) e a xira Haemulon chrysargyreum (Haemulidae), no
Arquipélago de Fernando de Noronha, Nordeste do Brasil. O saramunete e a xira são
semelhantes entre si, no padrão de coloração e no formato do corpo. Observamos que estas
duas espécies formam cardumes mistos regularmente, ao redor de recifes. Quando
perseguidos, pequenos grupos isolados de saramunetes se associam aos cardumes de xira e
se comportam de modo semelhante às xiras. As semelhanças de formato e coloração entre
as duas espécies provavelmente facilitam a formação de cardumes mistos e aumentam a
proteção contra predadores visualmente orientados, para ambas as espécies. Assim,
acreditamos que a associação protetora entre o saramunete e a xira pode ser considerada
como um tipo de “mimetismo social”, uma vez que nenhuma destas espécies é venenosa,
peçonhenta ou tem fortes estruturas mecânicas de defesa. Sugerimos, ainda, que exemplos
adicionais de mimetismo social possam envolver o saramunete e outras espécies listradas
de Haemulon.
Palavras chave: Mimetismo de proteção, mimetismo social, cardumes mistos,
Mulloidichthys martinicus, Haemulon chrysargyreum.
44
Introduction
Grunts (Haemulidae) are mostly nocturnal benthic feeders which during the day are
found in size-variable, inactive schools around reefs primarily for protection from incoming
open water predaceous fishes (Randall 1967, 1996). Grunts form a regular component of
shallow water reef fish communities in the tropical Western Atlantic(Randall 1967, Ehrlich
& Ehrlich 1972). The smallmouth grunt, Haemulon chrysargyreum, is a common reef
species in the Western Atlantic (Humann & DeLoach 2002) but restricted to oceanic
islands in Brazil (Rocha & Rosa 1999). It forages mostly at night and forms large inactive
schools around shallow reefs during the day (Randall 1967). The yellow goatfish,
Mulloidichthys martinicus (Mullidae), forages over sandy bottom both during day and night
(Randall 1967, 1996; Munro 1976). The goatfish also forms inactive schools around reefs
during the day, but generally with much fewer individuals than the grunt’s schools (pers
obs.).
Heterotypic schools of yellow goatfishes and grunts, mostly species of the genus
Haemulon, are widespread in the tropical Western Atlantic (Ehrlich & Ehrlich 1972). The
fishes within these mixed schools seem to gain protection against predators by increasing
the numbers of individuals that are similar in shape and colour pattern and thus have the
potential to confuse visually hunting fishes (Ehrlich & Ehrlich 1972). This mixed schooling
is called social mimicry (Moynihan 1968, Randall & McCosker 1993), synergic inviting
mimicry (Vane-Wright 1976), or school oriented mimicry (Dafni & Diamant 1984). In all
cases the similarity between the fish species that compose the mixed school probably
facilitates schooling and enhances the school cohesion. Although grunt-goatfish mixed
schools are a common sight in the tropical Western Atlantic, there are no reports on the
behaviour of these fishes during the group formation, and under which circumstances these
mixed groups are formed.
45
At the Fernando de Noronha Archipelago, off NE Brazil, the yellow goatfish and the
smallmouth grunt are regularly found in inactive mixed schools around the reefs. Herein we
report on the behaviour of the schools formed by the smallmouth grunt and the yellow
goatfish only, as well as mixed schools of both species. We suggest that the resemblance
and the association between these two fish species is an example of protective social
mimicry. Furthermore, we surmise that this form of protection may be widespread among
goatfishes and similarly looking, schooling other fish species anywhere.
Materials and methods
The study was conducted at the Fernando de Noronha Archipelago (03o50’ S; 32o25’
W), 340km off north-eastern Brazil, in June and July 2002 and June 2003. We observed,
video-recorded and photographed the behaviour of 10 schools of smallmouth grunt, 13
groups of yellow goatfish and 17 mixed schools of the two species, totalling 36 hours of
scuba-diving. Additionally, to simulate a chasing predator, we threatened (by swimming
straightforward towards the fishes) seven schools of about 10-50 yellow goatfish that were
inactive and sheltered in the reef, and three groups of about 4-10 ones that were foraging on
the sandy bottom in the open. We chased the groups to induce them to swim into open
water and recorded the behaviour of each group. During the study the depth ranged 4-15 m
and horizontal visibility ranged 10-30 m. Size of fishes is given as total length (TL)
estimates.
Results
Mulloidichthys martinicus and Haemulon chrysargyreum have a similar resemblance in
overall shape and colour pattern (Figure 1). At the study site the yellow goatfish formed
46
schools of about 4-50 individuals, and the grunt usually formed schools of about 40-1000
individuals. The monotypic inactive groups of yellow goatfishes were found sheltered
under ledges or in crevices in the reef, or were foraging on the sandy bottom in the open.
On the other hand, monotypic schools of the smallmouth grunt were always found inactive
in the water column in the open. The yellow goatfishes while in the water column in the
open were always mixed within the larger schools of inactive smallmouth grunts (Figure 1).
Within the mixed school, the yellow goatfish hovered in a head-down, oblique posture
similar to that displayed by the grunts within the school (Figure 2). The yellow goatfish and
smallmouth grunts that we recorded mixed together were of similar size class (15-20 cm
TL) and in only one small mixed school the yellow goatfish outnumbered (by about 10
individuals) the smallmouth grunt. Although the yellow goatfish appears to be a
conspicuous fish when solitary or in small groups, its colour pattern renders it
inconspicuous while schooling along with the smallmouth grunt (Figure 1). Experimentally
chased groups of yellow goatfish immediately joined the schools of the smallmouth grunt
when the former fish were induced to flee into open water (Figure 3).
47
Figure 1. The yellow goatfish (Mulloidichthys martinicus) is difficult to tell apart from the
smallmouth grunt (Haemulon chrysargyreum) while in mixed schools and viewed under
natural light.
Figure 2. The yellow goatfish adopts similar posture and behaviour while in a mixed
school with the smallmouth grunt. Note overall resemblance in shape and pattern between
the two species even under artificial light.
48
Figure 3. A previously inactive group of yellow goatfish (lower fish) joins a smallmouth
grunt school when chased by a would be predator (in this case a diver).
Discussion
We regard the association of the yellow goatfish with the smallmouth grunt as an
example of protective mimicry, so called social mimicry (Moynihan 1968, Randall &
McCosker 1993), synergic inviting mimicry (Vane-Wright 1976), and school oriented
mimicry (Dafni & Diamant 1984), since for all these categories a defensive function was
indicated. Our assumption is based in the fact that these two species are similar in shape
and colour pattern and that they form mixed schools. Additionally, when threatened, the
yellow goatfish always joined the larger schools of smallmouth grunt and behaved like
them. The case herein reported cannot be regarded as an example of Batesian or Müllerian
mimicry (Vane-Wright 1976), since the two fish species are neither venomous nor
poisonous, and do not appear to display any other feature which would deter or confound
potential predators (Dafni & Diamant 1984, Randall 1996, Froese & Pauly 2004).
This mixed schooling behaviour may be considered as advantageous for both the
smallmouth grunt and the yellow goatfish since their overall numbers increase while
49
mixing, and thus confuse visually hunting predators when targeting on its prey. However,
as the goatfish is generally less abundant than the smallmouth grunt (pers obs.), the
association much likely is more advantageous to the yellow goatfish than to the smallmouth
grunt. Moreover, the goatfish forage on sandy bottoms away from reef shelters and thus
they must rely on swimming off into open water as a form of defence against potential
predators (Gosline 1984). Therefore, we regard the large smallmouth grunt schools as very
important and predictable refuges for the yellow goatfish while fleeing in open water. We
predict that the yellow goatfish would join schools of other striped grunt species along the
Brazil’s coast, Haemulon squamipinna in Northeast Brazil and H. aurolineatum anywhere
on the coast being likely candidates for such mixed associations.
Since the genus Haemulon is restricted to the New World (both in the Atlantic and
Pacific), whereas Mulloidichthys is widespread in warm seas (Allen & Robertson 1994), it
would be of interest to check which additional genera and/or species these goatfishes
associate with. For instance, the Indo-Pacific Mulloidichthys mimicus joins schools of the
very similar lutjanid Lutjanus kasmira during the day (Allen et al. 2003, Froese & Pauly
2004). The yellow-striped goatfishes M. flavolineatus and M. manicolensis display a colour
pattern similar to that of the schooling caesionids Pterocaesio chrysozona and P. digramma
(Froese & Pauly 2004), also from the Indo-Pacific, and perhaps represent additional
instances of social, protective mimicry.
Acknowledgements
We thank two anonymous reviewers for greatly improving our manuscript; G. A. G. Pereira for
flying tickets; J. M. Silva-Jr (Centro Golfinho Rotador) for logistical and other help; the Atlantis
Diving Center for allowing free use of its facilities; S. R. Floeter, F. P. P. Leite, and V. S. Uieda
for suggestions on the manuscript; the Ibama for permits to study reef fishes at the Fernando de
50
Noronha National Marine Park; the CAPES, CNPq, FAPESP and PROAP-Unicamp for financial
support.
References
Allen, G.R. & Robertson, D.R. (1994). Fishes of the tropical eastern Pacific. Univ. Hawaii
Press. Honolulu.
Allen, G.R., Steene, R, Humann, P. & Deloach, N. (2003). Reef fish identification –
Tropical Pacific. New World Publications. Jacksonville.
Dafni, J. & Diamant, A. (1984). School-oriented mimicry, a new type of mimicry in fishes.
Mar. Ecol. Progr. Ser. 20: 45-50.
Ehrlich, P.R. & Ehrlich, A.H. (1972). Coevolution: heterotypic schooling in Caribbean reef
fishes. Am. Nat. 107: 157-160.
Froese, R. & Pauly, D. (Eds) (2004). FishBase. World Wide Web electronic publication.
www.fishbase.org, version 04/2004.
Gosline, W. A. (1984). Structure, function, and ecology in the goatfishes (family Mullidae).
Pacif. Sci. 38: 312-323.
Humann, P. & Deloach, N. (2002). Reef fish identification: Florida, Caribbean, Bahamas.
New World Publications. Jacksonville.
Moynihan, M. (1968). Social mimicry; character convergence versus character
displacement. Evolution 22 (2): 315-331.
Munro, J.L. (1976). Aspects of the biology and ecology of Caribbean reef fishes: Mullidae
(goatfishes). J. Fish Biol. 9: 79-97.
Randall, J.E. (1967). Food habits of reef fishes of the West Indies. Stud. Trop. Oceanogr.5:
665-847.
Randall, J.E. (1996). Caribbean reef fishes. 3rd. ed. T.F.H. Publications. Neptune City.
51
Randall, J.E. & Mccosker, J.E. (1993). Social mimicry in fishes. Revue Fr. Aquariol. 20:5-
8.
Rocha, L.A. & Rosa, I.L. (1999). A new species of Haemulon (Teleostei: Haemulidae)
from the Northeastern Brazilian coast. Copeia 1999(2): 447-452.
Vane-Wright, R.I. (1976). A unified classification of mimetic resemblances. Linn. Soc.
Lond 8 (1): 25-56.
52
DISCUSSÃO GERAL
Pseudupeneus maculatus pode ser caracterizada como uma espécie mais versátil e
generalista na escolha de substrato para forrageio que M. martinicus. Como outras espécies
de Pseudupeneus e Parupeneus (Gosline, 1984; McCormick, 1995), P. maculatus parece
ter potencial para forragear freqüentemente sobre áreas recifais tanto com cobertura de
areia e algas, como também sobre cobertura de algas pardas. Ainda, a menor freqüência
alimentar e o maior repertório de táticas alimentares de P. maculatus parecem estar
relacionados à inclusão de presas relativamente grandes e móveis em sua dieta, o que
reforçaria ainda mais a versatilidade desta espécie.
Pseudupeneus maculatus apresenta ampla variedade de táticas alimentares, sendo que
algumas delas, como perseguição e tocaia, aparentemente não estão registradas para
nenhuma outra espécie de Mullidae (q.v. Gosline, 1984; McCormick, 1995; Lukoschek &
McCormick, 2001). Foram ainda observados grupos de juvenis e adultos de P. maculatus
alimentando-se de plâncton, uma tática alimentar aparentemente rara e inesperada entre
espécies de Mullidae (Sierra et al., 1994, McCormick, 1995, Lukoschek & McCormick,
2001). A ampla distribuição e a grande abundância de P. maculatus em sua área de
ocorrência (Carvalho-Filho, 1999; Froese & Pauly, 2005) podem estar relacionadas à
versatilidade da espécie durante a alimentação e à sua capacidade de forragear sobre uma
ampla variedade de substratos.
O presente estudo indicou diferenças na seleção de substrato para forrageio de P.
maculatus quando comparado com estudos feitos no Caribe, onde a espécie parece
forragear principalmente sobre substrato arenoso (Itzkowitz, 1977; Aronson & Sanderson,
1987; J.P. Krajewski & R.M. Bonaldo obs. pess.). Particularmente na Jamaica, P.
maculatus parece alimentar-se exclusivamente sobre substrato arenoso (Itzkowitz, 1977).
No Caribe, os recifes parecem estar constituídos por, basicamente, de uma parte sólida
53
composta principalmente de corais, e áreas de substrato arenoso ao redor (J.P. Krajewski &
R.M. Bonaldo obs. pess.). Assim, o fato de P. maculatus aparentemente forragear somente
em substrato arenoso no Caribe deve estar relacionado à ausência de outros substratos em
que a espécie é capaz de forragear na região. Em recifes do Caribe onde há cobertura
abundante de algas, ou algas e areia, é esperado que P. maculatus também forrageie nestes
substratos e não fique restrito a áreas de substrato arenoso.
Mulloidichthys martinicus pode ser caracterizada como uma espécie menos versátil
que P. maculatus, forrageando principalmente sobre substrato arenoso. As táticas
alimentares apresentadas por M. martinicus são conhecidas para outras espécies de
Mullidade e, apesar de terem sido classificadas como quatro táticas distintas, apresentam
semelhança e derivam basicamente do comportamento de enterrar o focinho no substrato
(Gosline, 1984; McCormick, 1995; Lukoschek & McCormick, 2001). Assim, M.
martinicus parece ter impacto sobre a fauna associada a substrato arenoso, atingindo
profundidades de até cerca de 5cm (McCormick, 1995; Lukoschek & McCormick, 2001).
Como outras espécies do gênero, M. martinicus alimenta-se de presas relativamente
pequenas e sedentárias (e.g. Randall, 1967; Gosline, 1984), apresentando maior freqüência
de investidas no substrato que P. maculatus.
Mulídeos não apresentam defesas físicas e químicas contra predadores (Gosline,
1984; McCormick, 1995; Lukoschek & McCormick, 2001). Assim, acredita-se que o único
modo dos mulídeos escaparem de predadores seja fugir nadando rapidamente na coluna
d’água (McCormick, 1995; Lukoschek & McCormick, 2001). O comportamento alimentar
de algumas espécies de Mullidae de evitar enterrar o focinho até a altura do olho em
substrato arenoso e forragear em grupos parece estar relacionado a uma maior detecção e
proteção contra predadores (Lukoschek & McCormick, 2001). No presente estudo, é
relatado um possível caso de mimetismo de proteção entre M. martinicus e H.
54
chrysargyreum (Haemulidae). Mulloidichthys martinicus aproveita os grandes cardumes de
H chrysargyreum, abundantes na área de estudo, para confundir e evitar predadores
potenciais. Este tipo de defesa parece ser especialmente importante para M. martinicus, que
forrageia em áreas abertas e apresenta atividade alimentar tanto diurna como noturna
(Randall, 1967; Munro, 1976). Assim, o mimetismo de proteção é usado por indivíduos de
M. martinicus que estão inativos durante o dia e se abrigam entre os cardumes de H
chrysargyreum e por indivíduos que estão forrageando e, ao se sentirem ameaçados,
buscam momentaneamente a proteção entre os cardumes.
Mulloidichthys martinicus e P. maculatus diferiram em todos os aspectos estudados
de seu comportamento e atividade alimentar. Estes resultados, juntamente com resultados
de estudos recentes sobre outras espécies de Mullidae (e.g. McCormick, 1995; Lukoschek
& McCormick, 2001), indicam que mulídeos não são simplesmente fossadores generalistas
de substrato não consolidado. Espécies de Mullidae podem forragear sobre uma grande
variedade de substratos do recife, tendo impacto também na fauna associada a algas.
REFERÊNCIAS
Aronson, R. B. &. Sanderson S. L (1987). Benefits of heterospecific foraging by the
Caribbean wrasse, Halichoeres garnoti (Pisces: Labridae). Environmental Biology of
Fishhes 18, 303-308.
Carvalho-Filho, A. (1999). Peixes: Costa Brasileira. 3a edição. Editora Melro, São Paulo.
283pp.
Froese, R. & Pauly D. (Eds.) (2005). FishBase.World Wide Web electronic publication.
www.fishbase.org.
Gosline, W. A. (1984). Structure, function, and ecology in the goatfishes (family Mullidae).
Pacific Science 38, 312-323.
55
Itzkowitz, M. (1977). Social dynamics of mixed-species groups of Jamaican reef fishes.
Behavioural Ecology and Sociobiology 2, 361-384.
Lukoschek, V. & McCormick, M. I. (2001). Ontogeny of diet changes in a tropical benthic
carnivorous fish, Parupeneus barberinus (Mullidae): Relationship between foraging
behaviour, habitat use, jaw size, and prey selection. Marine Biology 128, 1099-1113.
McCormick, M. I. (1995). Fish feeding on mobile benthic invertebrates: influence of
variability in habitat associations. Marine Biology 121, 627-637.
Munro, J.L. (1976). Aspects of the biology and ecology of Caribbean reef fishes: Mullidae
(goat fishes). Journal of Fish Biology 9, 79-97.
Randall, J. E. (1967). Food habits of reef fishes of the West Indies. Studies in Tropical
Oceanography Miami 5, 665-847.
Sierra, L. M., Claro, R. & Popova, O. A. (1994). Alimentación y relaciones tróficas. In
Ecología de los peces marinos de Cuba (Claro, R., ed.), pp. 263-320. Quintana Roo,
México: Centro de Investigaciones de Quintana Roo.